Hardware device temperature control with expected lifetime calculation

Abstract

Embodiments herein describe coupling traditional fan and shaper control along with aggregated knowledge of the temperature history of a hardware device to optimally manage the temperature of the hardware device to preserve its expected life while also providing the lower power, best performing solution possible. In one embodiment, a cooling application manages the expected life by trading off performance and power versus temperature to achieve a desired (or accepted) lifetime. In one embodiment, the cooling application calculates a historical temperature value for the hardware device which is then used to determine the expected life of the hardware device.

We claim: 
     
       1. A method, comprising:
 determining an expected life of a hardware device in a computing system based on a plurality of historical temperature measurements and a current temperature measurement; 
 comparing the expected life to a desired remaining life of the hardware device, wherein the desired remaining life is based on a desired lifespan and a total runtime of the hardware device; 
 upon determining the expected life is less than the desired remaining life, decreasing the temperature of the hardware device by changing an operating parameter of a cooling element in the computing system; and 
 before increasing the temperature of the hardware device, determining that a timeout value has expired, wherein the timeout value prevents a cooling application from increasing the temperature for a predefined period of time. 
 
     
     
       2. The method of  claim 1 , wherein the expected life is an estimate of time remaining before the hardware device is predicted to fail. 
     
     
       3. The method of  claim 1 , further comprising:
 determining the desired remaining life by subtracting the total runtime from the desired lifespan, wherein the desired lifespan is a fixed value that is set by a system administrator. 
 
     
     
       4. The method of  claim 1 , further comprising, after decreasing the temperature of the hardware device:
 determining an updated value of the expected life using the plurality of historical temperature measurements and an updated current temperature measurement; and 
 upon determining the updated value of the expected life is greater than the desired remaining life, increasing the temperature of the hardware device by changing the operating parameter of the cooling element, wherein the operating parameter adjusts an amount of cooling performed by the cooling element. 
 
     
     
       5. The method of  claim 1 , further comprising, before increasing the temperature of the hardware device:
 upon determining that the timeout value has not expired, maintaining the parameter of the cooling element at a current value set when decreasing the temperature of the hardware device. 
 
     
     
       6. The method of  claim 1 , wherein determining the expected life of the hardware device comprises:
 determining an effective temperature of the hardware device based on the Arrhenius equation and the plurality of historical temperature measurements. 
 
     
     
       7. The method of  claim 1 , wherein determining the expected life of the hardware device comprises:
 determining an effective temperature of the hardware device using a technique that predicts failure caused by at least one of metal electro-migration, negative-bias temperature instability (NBTI), and hot-carrier injection (HCI). 
 
     
     
       8. A non-transitory computer readable medium having program instructions embodied therewith, the program instructions executable by a processor to perform an operation, the operation comprising:
 determining an expected life of a hardware device in a computing system based on a plurality of historical temperature measurements and a current temperature measurement; 
 comparing the expected life to a desired remaining life of the hardware device, wherein the desired remaining life is based on a desired lifespan and a total runtime of the hardware device; 
 upon determining the expected life is less than the desired remaining life, decreasing the temperature of the hardware device by changing an operating parameter of a cooling element in the computing system; and 
 determining the desired remaining life by subtracting the total runtime from the desired lifespan, wherein the desired lifespan is a fixed value that is set by a system administrator. 
 
     
     
       9. The non-transitory computer readable medium of  claim 8 , wherein the expected life is an estimate of time remaining before the hardware device is predicted to fail. 
     
     
       10. The non-transitory computer readable medium of  claim 8 , wherein the operation further comprises, after decreasing the temperature of the hardware device:
 determining an updated value of the expected life using the plurality of historical temperature measurements and an updated current temperature measurement; and 
 upon determining the updated value of the expected life is greater than the desired remaining life, increasing the temperature of the hardware device by changing the operating parameter of the cooling element. 
 
     
     
       11. The non-transitory computer readable medium of  claim 10 , wherein the operation further comprises, before increasing the temperature of the hardware device:
 determining that a timeout value has expired, wherein the timeout value prevents a cooling application from increasing the temperature for a predefined period of time. 
 
     
     
       12. The non-transitory computer readable medium of  claim 11 , wherein the operation further comprises, before increasing the temperature of the hardware device:
 upon determining that the timeout value has not expired, maintaining the operating parameter of the cooling element at a current value set when decreasing the temperature of the hardware device. 
 
     
     
       13. A computing system, comprising:
 a hardware device; and 
 a cooling application configured to:
 determine an expected life of the hardware device in the computing system based on a plurality of historical temperature measurements and a current temperature measurement; 
 compare the expected life to a desired remaining life of the hardware device, wherein the desired remaining life is based on a desired lifespan and a total runtime of the hardware device; 
 upon determining the expected life is less than the desired remaining life, decrease the temperature of the hardware device by changing an operating parameter of a cooling element in the computing system; and 
 determine the desired remaining life by subtracting the total runtime from the desired lifespan, wherein the desired lifespan is a fixed value that is set by a system administrator. 
 
 
     
     
       14. The computing system of  claim 13 , wherein the expected life is an estimate of time remaining before the hardware device is predicted to fail. 
     
     
       15. The computing system of  claim 13 , wherein the cooling application is configured to, after decreasing the temperature of the hardware device:
 determine an updated value of the expected life using the plurality of historical temperature measurements and an updated current temperature measurement; and 
 upon determining the updated value of the expected life is greater than the desired remaining life, increase the temperature of the hardware device by changing the operating parameter of the cooling element. 
 
     
     
       16. The computing system of  claim 15 , wherein the cooling application is configured to, before increasing the temperature of the hardware device:
 determine that a timeout value has expired, wherein the timeout value prevents the cooling application from increasing the temperature for a predefined period of time. 
 
     
     
       17. The computing system of  claim 16 , wherein the cooling application is configured to, before increasing the temperature of the hardware device:
 upon determining that the timeout value has not expired, maintain the operating parameter of the cooling element at a current value set when decreasing the temperature of the hardware device.